How do we select a site to install sensors

Criteria:

Experiences saltwater intrusion (near coast)

Blackwater system (manually subset coastal systems)

Experiences a seasonal change in color (green/[blue + red])

Accessible (bridges to install sensors or could use a boat)

Initial Analysis Methods

Manually selected rivers in NC

Selected rivers that I thought would be large enough and have high DOC contractions

Calculated distance from river mouth

Pulled landsat data

Using the Color of Rivers code, pulled landsat pixels that intersected NHD flowlines

Used the Dynamic Surface Water Extent (DSWE) algorithm, which casts pixels as not water (0), water high confidence (1), water low confidence (2), vegetated water high confidence (3), vegetated water low confidence (4) to remove non-water pixels.

Removed pixels with a DSWE of 0 and filtered for pixels with a mean DSWE less than 1.8. Chose this value from visual inspection on the intracoastal waterways.

Mean DSWE after 0s were removed for whole time series

After pixels are removed

Calculated spectral indices

Focused on the index that was correlated with salinity in St. Johns analysis (Green/(Blue+Red)). Calling the Saltwater Intrusion Index (SWII) for now

Time series of the SWII from Landsat 5,7, and 8 for the intracoastal waterway, colored by distance from the southern most outlet (Pungo River)

Zoomin on 2020 & 2021

Takeaways

I am not sure what remote sensing information to consider for this analysis. The Green/(Blue+Red) was correlated with salinity and DOC in the St. Johns but this is also the visual spectrum and influenced by a lot of factors. But using the Green/(Blue+Red) index, it seems there is seasonal variation in many coastal systems in NC.

Many rivers see an expected seasonal changes in the Green/(Blue+Red) index and higher values downstream (Perquimans, Pasquotank, and New). But some rivers show higher Green/(Blue+Red) values at the most upstream location (Scuppernong). This probably indicates cDOM is not solely driving the color of these waters. Algae, emergent vegetation, turbidity, or other things may be driving the color.

Scuppernong River

Visible pixels

Time series

Zoomin on 2018 & 2019

Look at anomaly between the most down stream locations

Milltail Creek

Visible pixels

Time series

Zoomin on 2017 & 2018

Look at anomaly between the most down stream locations

New River

Visible pixels

Time series

Zoomin on 2008 to 2010

Look at anomaly between the most down stream locations

Pasquotank River

Visible pixels

Time series

Zoomin on 2007 to 2010

Look at anomaly between the most down stream locations

Perquimans River

Visible pixels

Time series

Zoomin on 2007 & 2010

Look at anomaly between the most down stream locations

White Oak

Visible pixels

Time series

Zoomin on 2007 to 2010

Look at anomaly between the most down stream locations